Sökning: "prestanda i realtid"

Visar resultat 1 - 5 av 124 uppsatser innehållade orden prestanda i realtid.

  1. 1. Towards Adaptive Image Resolution for Visual SLAM on Resource-constrained Devices

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Herman Blenneros; [2023]
    Nyckelord :Visual localization and mapping; Runtime-control; Resource-constrained devices; Bildbaserad lokalisering och kartläggning; Realtidsreglering; Resursbegränsade enheter;

    Sammanfattning : Today, a large number of devices with small form factors and limited resources are being integrated with processes to perform complex tasks such as localization and mapping. One example of this are headsets used for Extended Reality. LÄS MER

  3. 2. Comparison between Smoothed-Particle Hydrodynamics and Position Based Dynamics for real-time water simulation

    Kandidat-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Rasmus Andersson; Erica Tjernell; [2023]
    Nyckelord :Fluid simulation; Lagrangian fluid simulation; Smoothed-Particle Hydronamics; Position Based Dynamics; Vätskesimulering; Lagransk vätskesimulering; Smoothed-Particle Hydronamics; Position Based Dynamics;

    Sammanfattning : Two of the methods common in video game fluid simulation are SmoothedParticle Hydrodynamics (SPH), and Position Based Dynamics (PBD). They are both Lagrangian methods of fluid simulation. SPH has been used for many years in offline simulations and has truthful visuals, but is not as stable as the newer method PBD when using larger timesteps. LÄS MER

  4. 3. Predictive Controllers for Load Transportation in Microgravity Environments

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Sujet Phodapol; [2023]
    Nyckelord :Model Predictive Control; Decentralized control; Multi-robot systems; Space technology; Modell-prediktiv reglering; Decentraliserad reglering; Multirobotsystem; Rymdteknologi;

    Sammanfattning : Space activities have been increasing dramatically in the past decades. As a result, the number of space debris has also increased significantly. Therefore, it is necessary to clean up and remove them to prevent a collision between space debris and spacecraft. LÄS MER

  5. 4. Exploration and Evaluation of RNN Models on Low-Resource Embedded Devices for Human Activity Recognition

    Master-uppsats, KTH/Mekatronik och inbyggda styrsystem

    Författare :Helgi Hrafn Björnsson; Jón Kaldal; [2023]
    Nyckelord :Recurrent Neural Networks; Long Short-Term Memory Networks; Embedded Systems; Human Activity Recognition; Edge AI; TensorFlow Lite Micro; Recurrent Neural Networks; Long Short-Term Memory Networks; Innbyggda systyem; Mänsklig aktivitetsigenkänning; Edge AI; TensorFlow Lite Micro;

    Sammanfattning : Human activity data is typically represented as time series data, and RNNs, often with LSTM cells, are commonly used for recognition in this field. However, RNNs and LSTM-RNNs are often too resource-intensive for real-time applications on resource constrained devices, making them unsuitable. LÄS MER

  7. 5. Smart Tracking for Edge-assisted Object Detection : Deep Reinforcement Learning for Multi-objective Optimization of Tracking-based Detection Process

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Shihang Zhou; [2023]
    Nyckelord :Tracking-By-Detection; Deep Reinforcement Learning; Multi-Objective Optimization; Spårning genom detektion; Djup förstärkningsinlärning; Multiobjektiv optimering;

    Sammanfattning : Detecting generic objects is one important sensing task for applications that need to understand the environment, for example eXtended Reality (XR), drone navigation etc. However, Object Detection algorithms are particularly computationally heavy for real-time video analysis on resource-constrained mobile devices. LÄS MER